2

i try to do the following:

  • Download a PDF file to memory
  • Convert the first page to an image
  • Use that image with tweepy

I tried the following code, but run into an error. 

from PIL import Image
from pdf2image import convert_from_path
from urllib.request import urlopen
from io import StringIO, BytesIO

url = 'http://somedomain.com/assets/applets/internet.pdf'
scrape = urlopen(url) # for external files
pdfFile = BytesIO(scrape.read())
pdfFile.seek(0)
pages = convert_from_path(pdfFile,last_page=1, dpi=100)

for page in pages:
    page.save('/home/out.jpg', 'JPEG')

Here is the error:

TypeError: Can't convert '_io.BytesIO' object to str implicitly

The generated image should later be used to upload it to twitter by tweepy. I don't need to store it to disk, that's why i try to do all in memory. Anybody who could help me please?

Lionking
  • 53
  • 9
  • You need to use `convert_from_bytes` method instead of `convert_from_path` – kip Jun 08 '18 at 15:44
  • @kip I changed the code to pages = convert_from_bytes(pdfFile,last_page=1, dpi=100) Also imported this function, but still get an error: TypeError: a bytes-like object is required, not '_io.BytesIO' – Lionking Jun 08 '18 at 21:54
  • 1
    try pass to above method a `bytes` from the `BytesIO` object, maybe with `getvalue()`, something like: `convert_from_bytes(pdfFile.getvalue()`, but I think that `scrape.read()` return a `bytes` object.... `convert_from_bytes(scrape.read()` – kip Jun 09 '18 at 03:29
  • 2
    To use scrape.read() directly did the trick. Thank you a lot. – Lionking Jun 09 '18 at 09:25

0 Answers0